Pure Player - Page 49 - diyAudio
Go Back   Home > Forums > Source & Line > PC Based

PC Based Computer music servers, crossovers, and equalization

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 23rd August 2013, 07:09 AM   #481
diyAudio Member
 
Join Date: Mar 2011
Quote:
Originally Posted by phofman View Post
Jack by principle cannot improve sound quality, it is just a layer between application and the sound layer - alsa, used for transporting data and control between audio production applications. It introduces additional conversions to float32, its internal format, adds to your CPU load. Audiophile linux is created by a guy who honestly states has no background in linux audio, just did what heard in internet discussions.

Also low latency adds to your machine CPU load. But is it audible? I very much doubt that and noone has made/reported about a proper blind listening test with positive results.
I have neither special backroung in linux audio, nor tried another distribution in my system. I know (it is easily audible) that:

1) lowering the latency in both windows and linux improves the sound
2) after setting the same latency Audiophile linux sounds better than Windows XP.
3) Jack sounds better than alsa and pulse. This is possibly due to the ability to adjust latency.

Is there another way besides Jack to lock exclusively the sound device, playback in realtime process priority and -most important- adjust the latency?

Do you have in mind a better audio oriented distribution? I would be happy to try it.

Last edited by npetralias; 23rd August 2013 at 07:11 AM.
  Reply With Quote
Old 23rd August 2013, 10:15 AM   #482
phofman is offline phofman  Czech Republic
diyAudio Member
 
Join Date: Apr 2005
Location: Pilsen
Quote:
Originally Posted by npetralias View Post
I know (it is easily audible) that:

1) lowering the latency in both windows and linux improves the sound
2) after setting the same latency Audiophile linux sounds better than Windows XP.
3) Jack sounds better than alsa and pulse. This is possibly due to the ability to adjust latency.
Unless you did a blind listening test I do not see how you could "know" it. You feel it sounds that way, having read many also subjective reports as well. I have also read a few subjective reports prefering the larger latency. All observations being subjective, there is no point in discussing technical causes.

My blind test on my gear revealed I was not able to tell any difference between low and large latency. How will your test end up? E.g. you can use my rudimentary script Tool for A/B Testing | Blog IVITERA a.s.


Quote:
Originally Posted by npetralias View Post
Is there another way besides Jack to lock exclusively the sound device, playback in realtime process priority and -most important- adjust the latency?
Jack adjusts latency via alsa. It is just a layer above it. Most decent playback applications can adjust buffer size in its alsa output module. Locking sound device is just a question of which alsa output device string you use. Jack just shields you away from the lower-level mechanism and hides the actual principles from you. IMO you being a technical guy want to get down to the core and learn the real stuff, not just a subset of features jack developers provided for DAW users/pro's.


Quote:
Do you have in mind a better audio oriented distribution? I would be happy to try it.
It really depends what you are looking for. General-purpose PC, streamlined PC for audio only, embedded headless device - many choices. They all need some linux knowledge to setup reasonably, nothing complicated, just not being afraid of command line (which I assume you as a developer of command-line pureplay are not ).
  Reply With Quote
Old 23rd August 2013, 10:53 AM   #483
diyAudio Member
 
Join Date: Mar 2011
Quote:
Originally Posted by phofman View Post
Unless you did a blind listening test I do not see how you could "know" it. You feel it sounds that way, having read many also subjective reports as well. I have also read a few subjective reports prefering the larger latency. All observations being subjective, there is no point in discussing technical causes.

My blind test on my gear revealed I was not able to tell any difference between low and large latency. How will your test end up? E.g. you can use my rudimentary script Tool for A/B Testing | Blog IVITERA a.s.
Each one has it's own methods of auditioning a system or component.
After many years, knowing my system very well i tend to listen better with eyes open but i performed some blind tests with friends and all 3 of them preffered audiophile linux with jack and latency as low as possible. The key word here is "prefer", because taste varies. So i like the fast, raw, attacking low latency setting rather than the safe, sweet-ish, fluffy and lazy normal one. Perhaps it's all a mater of taste.

Quote:
Originally Posted by phofman View Post
Jack adjusts latency via alsa. It is just a layer above it. Most decent playback applications can adjust buffer size in its alsa output module. Locking sound device is just a question of which alsa output device string you use. Jack just shields you away from the lower-level mechanism and hides the actual principles from you. IMO you being a technical guy want to get down to the core and learn the real stuff, not just a subset of features jack developers provided for DAW users/pro's.




It really depends what you are looking for. General-purpose PC, streamlined PC for audio only, embedded headless device - many choices. They all need some linux knowledge to setup reasonably, nothing complicated, just not being afraid of command line (which I assume you as a developer of command-line pureplay are not ).
Can you suggest me a player or another direct way to adjust latency? Unfortunately i don't have enough time to "go deep" right now...


ps. your tool looks great!
  Reply With Quote
Old 23rd August 2013, 10:41 PM   #484
phofman is offline phofman  Czech Republic
diyAudio Member
 
Join Date: Apr 2005
Location: Pilsen
I have been at this countless times, never ever anyone claiming to hear the low latency effect dared to take the blind test. I did on my gear, with results well below confidence threshold for my ears.

If you want to (which I very much doubt), you can take the test using that script.

Quote:
Can you suggest me a player or another direct way to adjust latency
Any direct alsa player which lets you setup buffer and/or period size, such as mpd (parameter period_time). For basic players for testing purpose (e.g. the blind test) the easiest tool is aplay with parameters --period-time and --buffer-time. If you want to, we can talk about what these actually do so that you see what this low latency thing actually does in the system. Most of this is hidden from you in windows.
  Reply With Quote
Old 26th August 2013, 11:11 AM   #485
diyAudio Member
 
Join Date: Mar 2011
I did the blind test with a help of a friend and it was easy to tell each time which setting is A (10ms latency) and B (2ms latency). I 'm afraid though that the matter is more complicated than concentrating on latency. In both Linux and Windows there are different results with different combinations of buffer size, latency and sample rate. The "key" IMHO is not latency itself but the right combination of the above, although as a general rule things are better as latency is lowered. This may be also dependant on the DAC or USB(firewire) interface used. Moreover, most of the players and plugins sound differently.

In Audiophile Linux IMO Deadbeef is the best, Audacious and gmusicbrowser just good, alsaplayer and musique not so good... Regarding output plugins: Jack is very good, Pulse good, Alsa not so good. So Deadbeef+Jack+low latency is the best for my ears-system but can anyone explain why?

@phofman: Have you tried the above combinations? Is the "audible" playback result similar to you?

Last edited by npetralias; 26th August 2013 at 11:14 AM.
  Reply With Quote
Old 26th August 2013, 12:25 PM   #486
phofman is offline phofman  Czech Republic
diyAudio Member
 
Join Date: Apr 2005
Location: Pilsen
Quote:
Originally Posted by npetralias View Post
I did the blind test with a help of a friend and it was easy to tell each time which setting is A (10ms latency) and B (2ms latency).

That is very interesting, please yould you provide more details about the test? What command you tested, how the test was administered, what results. Thanks a lot


Quote:
Regarding output plugins: Jack is very good, Pulse good, Alsa not so good.
So basically you say that going through jack into alsa sounds better than going into alsa directly. Do you see where my doubts come from? Jack is just an additional layer with multiple sample conversions.

Quote:
So Deadbeef+Jack+low latency is the best for my ears-system but can anyone explain why?
I will not delve into technical explanation of subjective feelings.

Quote:
@phofman: Have you tried the above combinations? Is the "audible" playback result similar to you?
The combination itself says nothing about the actual setup - output plugins, configuration of the output plugins, etc.... There can be countless differences. Plus "sounds better" is extremely individual - everyone prefers something else. The only relatively objectively comparable result is "showing a difference can be identified/heard".
  Reply With Quote
Old 26th August 2013, 01:49 PM   #487
diyAudio Member
 
Join Date: Mar 2011
Quote:
Originally Posted by phofman View Post
That is very interesting, please yould you provide more details about the test? What command you tested, how the test was administered, what results. Thanks a lot
No commands, just changing parameters in jack or pureplayer and playing 15 seconds of the song randomly. It is very easy to guess the settings each time.

Quote:
Originally Posted by phofman View Post
So basically you say that going through jack into alsa sounds better than going into alsa directly. Do you see where my doubts come from? Jack is just an additional layer with multiple sample conversions.
Yes. Obviously if players could utilize alsa the way jack does the result would be better without using jack.

Quote:
Originally Posted by phofman View Post
The combination itself says nothing about the actual setup - output plugins, configuration of the output plugins, etc.... There can be countless differences. Plus "sounds better" is extremely individual - everyone prefers something else. The only relatively objectively comparable result is "showing a difference can be identified/heard".
So you cannot tell me if you hear any difference with any combination.

Quote:
Originally Posted by phofman View Post
I will not delve into technical explanation of subjective feelings.
I am very sorry, but this is our hobby and if we cancel our hearing there is really nothing to discuss here. Its subjectiveness may be its beauty. Excuse me but i would not like to go any further on this conversation.
  Reply With Quote
Old 26th August 2013, 02:49 PM   #488
phofman is offline phofman  Czech Republic
diyAudio Member
 
Join Date: Apr 2005
Location: Pilsen
Quote:
Originally Posted by npetralias View Post
No commands, just changing parameters in jack or pureplayer and playing 15 seconds of the song randomly. It is very easy to guess the settings each time.
I do not belive distinguishing between 2ms and 10ms latencies is "easy" unless something is fundamentally wrong in your system (e.g. audible dropouts in the 2ms setup, interrupts causing audible noise in your system).



Quote:
Yes. Obviously if players could utilize alsa the way jack does the result would be better without using jack.
I see nothing obvious in this. In addition, the player itself plays the same, whatever latency in jack you setup, only the writing thread (producer) is awoken by jack more often, just like the driver (CPU) is called by the soundcard interrupt more often. The playback application thread and the soundcard DMA transfer run in the same pace, whatever the latency setup in between.

Of course you can setup alsa playback for the same latency like jack, in the end it is always alsa handling the data transfer into the soundcard's DMA region in RAM. Jack itself does not reduce any latency at all (how could it), it is just a plumbing infrastructure.


Quote:
So you cannot tell me if you hear any difference with any combination.
My ears neither my wife's did not hear any difference in latency on my setup in a properly administered blind test (i.e. we did not know what alternative was being played). Our correct results fell below the threshold of being statistically significant to become more than random guesses.

Quote:
I am very sorry, but this is our hobby and if we cancel our hearing there is really nothing to discuss here. Its subjectiveness may be its beauty. Excuse me but i would not like to go any further on this conversation.
Discussing technical causes for subjective feelings is definitely not something I want to do. Good luck.
  Reply With Quote
Old 26th August 2013, 03:07 PM   #489
Wombat is offline Wombat  Germany
diyAudio Member
 
Wombat's Avatar
 
Join Date: Jul 2002
Location: Germany
That is one problem that is coming up when asking for blind tests. The informed audiophile just will add to his signature: "All subjective claims are blind tested" and done. No interest in doing this consequent.
I donīt want to smaller the work of npetralias but imagine a double-blind-test shows no result? All those years of work for nothing? The fan crowd, the fun...
__________________
If time permits - stuff some parsley in your ears and listen how it grows!
  Reply With Quote
Old 31st August 2013, 05:10 PM   #490
JaeK is offline JaeK  Sweden
diyAudio Member
 
Join Date: Aug 2013
This thread is hilarious.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Pure digital crossover cka3o4nuk Digital Line Level 14 12th September 2010 12:08 AM
Pure CD Player and DVD Transport Mechanism inside - what about this ?? tiefbassuebertr Digital Source 2 23rd February 2010 09:36 PM
Eminence Pure rossco Car Audio 2 28th March 2004 11:40 PM
Pure Delight Alain Dupont Pass Labs 107 8th February 2004 03:48 PM


New To Site? Need Help?

All times are GMT. The time now is 01:57 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright Đ1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2